Cell morphology and property relationships of microcellular foamed PVC/wood-fiber composites.
Cellulose materials, such as wood-fibers, combined with synthetic polymers are known as plastic/wood-fiber composites. The main reason for combining wood-fibers with plastic is to merge the desirable properties of each component, thereby creating a new material with extended utility (1-3). Although these composites are known to be superior to the neat polymers in terms of material cost and stiffness, their strength performance is generally lower. The decreased strength is likely a result of the natural incompatibility of phases during the mixing of the hydrophilic wood-fiber with the hydrophobic polymeric matrix (3-9). A phase incompatibility will yield very weak interactions and thus a weak interface between the fiber and the matrix. Moreover, strong fiber-fiber interactions resulting from hydrogen bonding limit the dispersion of the fibers in the matrix. These fundamental problems have been examined by a number of investigators (3-9). Surface modification of the fibers by coupling or compatibilizing agents, which facilitate the fiber dispersion and induce bond formation between the polymer and the fibers, is commonly suggested as an effective solution to these problems.
In general, the lower material cost and improved stiffness of wood-fiber composites, due to the filler reinforcement effect, are accomplished at the expense of other properties such as the ductility and the impact strength (7-9). Although the plastic/wood-fiber composites have been commercialized and are competing with wood products and certain plastics, the usefulness of plastic/wood-fiber composites has been limited because of their brittleness, low impact strength, and high density compared to unfilled plastics and natural wood.
A microcellular polymer is a foamed plastic characterized by a cell density in the range of [10.sup.9] to [10.sup.15] cells/[cm.sup.3] and fully grown cells in the range of 0.1 to 10 [[micro]meter]. In the last decade, interest in the production of these materials has grown for several reasons. In particular, they offer the benefits of the reduced material usage and lowered weight while enhancing the impact strength (10-16), toughness (16, 17), fatigue life (18, 19), and thermal stability (20). These improvements are due to the presence of cells that are much smaller than those of conventional foams. These small bubbles inhibit crack propagation by blunting the crack tip and increasing the amount of energy needed to propagate the crack (12, 13). The presence of these cells can also reduce the specific density of the polymer in the range of 75% or higher (12, 21-23).
Since microcellular foamed plastics are known to exhibit a higher impact strength, higher toughness, increased fatigue life, and enhanced thermal stability, microcellular foaming of wood-fiber composites would improve these properties in the composites as well. Especially, it is expected that the deteriorated impact strength of plastic/wood-fiber composites will be improved significantly. Therefore, microcellular foaming of wood-fiber composites would lead to the creation of a new class of materials with enhanced properties and characteristics. We believe that the development of microcellular structure in these polymer/wood-fiber composites will significantly increase their potential of industrial application.
Microcellular foaming of PVC materials has been extensively studied (12, 22, 24-26). The processing-structure (22, 24, 25) and structure-property (12, 26) relationships of microcellular foamed PVC have been derived. However, the property characterization of microcellular foamed PVC/wood-fiber composites has not been conducted extensively.
The main motivation of this study is to investigate the effects of microcellular structure developed in PVC/wood-fiber composites on the mechanical and physical properties. Our recent investigations have successfully demonstrated the feasibility of developing and tailoring the microcellular structures in PVC/wood-fiber composites (22, 23). Since the introduction of microcellular structure will improve impact strength and reduce weight as mentioned above, it is expected that microcellular foamed plastic/wood-fiber composites will offer better performance in these properties. However, it is also expected that the development of a foamed structure will deteriorate some mechanical properties such as the tensile strength and modulus. In this context, it is the purpose of this study to quantitatively characterize the impact strength, tensile strength, and modulus of microcellular PVC/woodfiber composites. The effect of cell morphology on the tensile and impact properties of microcellular foamed PVC/wood-fiber composites, and the effect of fiber surface treatment on the properties of microcellular foamed composites are central to this study.
Microcellular foamed PVC/wood-fiber composites were produced in four steps. First, the composites were manufactured by compounding PVC with either untreated or treated wood-fibers in a high intensity turbine mixer (Werner-Pfleiderer Gelimat G-1). Second, the compounded materials were compression-molded into panels using a hot press. Third, sorption experiments were carried out to determine the sorption parameters for the samples. Finally, the samples were microcellular foamed by saturating the samples with gas followed by rapidly dropping the gas solubility. The cell morphology of foamed samples was characterized by a scanning electron microscope (SEM). Tensile and impact tests were performed on the foamed composites to investigate the dependence of these properties on the cell morphology of foamed composites.
The polymer matrix used in this study was PVC (Royal Plastics Ltd., Geon 103EPF76, K value 66). A commercial type of wood-fiber from Interfibe Corporation was used as a filler. The fiber length and diameter were in the range of 30-40 [[micro]meter] and 0.3-1.0 [[micro]meter], respectively. Gamma-aminopropyltriethoxysilane (Union Carbide Corporation, A-1100) (27) and dioctyl phthalate (DOP) were used as a coupling agent and a plasticizer, respectively. C[O.sub.2] (Matheson Gas Products, commercial 99.5% min) was utilized as a foaming agent. All chemicals received from the manufacturers remained unmodified.
Manufacture of the Composites
The composites were manufactured by mixing PVC with either untreated or silane-treated fibers in a high intensity turbine mixer. The compounding conditions and the treatment of wood-fibers with silane A-1100 are described in detail elsewhere (3, 27). The woodfiber content and the plasticizer concentration used in the composites were fixed to be 30 phr (parts per hundred resin) and 20 phr, respectively.
These compounded mixtures were compression-molded into panels (1.5 mm in thickness) in a hydraulic press (Wabash Metal Products Inc., Model 50-1818-2tm) at 190 [degrees] C for 4 minutes using 30 tons of pressure. From these panels, the tensile dogbone-shaped specimens (Type IV tensile specimens of an ASTM standard D638) were made for foaming and tensile tests. Rectangular samples of 12.7 mm x 63.5 mm were cut from the panel sheets for impact tests. After foaming, the impact test specimens were machined to their original size.
Microcellular Foaming Experiments
Microcellular foaming experiments were conducted as follows: PVC and wood-fiber composites/C[O.sub.2] solutions were formed using a high pressure gas saturation procedure outlined previously (23). The saturation time and the pressure were 2 days and 5.51 MPa (800 psi), respectively. The samples were weighed before and after the C[O.sub.2] saturation to determine the weight gain of C[O.sub.2] in the samples. The measured maximum amount of C[O.sub.2] absorbed, when the pressure was released, was taken as the "measured" concentration of gas in the polymer or composite. After saturation, the C[O.sub.2]-saturated samples were foamed via a rapid solubility drop involving a combined pressure decrease and temperature increase. This was achieved by taking the samples out of the pressure chamber and dipping them for 5 seconds in a hot glycerol bath which was maintained at various temperatures (70 [degrees] C, 90 [degrees] C, 110 [degrees] C, and 130 [degrees] C).
In previous investigations, the foaming time and temperature have been identified as effective process parameters for controlling the cell morphologies of foamed PVC and wood-fiber composites (22). By tailoring these process parameters, a set of desired void fractions, cell-population densities, and average cell sizes were achieved so that the mechanical properties of foamed PVC and wood-fiber composites could be characterized in terms of these structural parameters.
Characterization of Foams
The densities of PVC and wood-fiber composites were measured by a water displacement technique (ASTM standard D792). First, the weights of samples were measured in air ([M.sub.a]) and while immersed in distilled water ([M.sub.w]). Then, the densities of the unfoamed ([Rho]) and foamed ([[Rho].sub.F]) materials were determined as
Density = 0.9975 ([M.sub.w]/[M.sub.w]) (1)
Since Collias and Baird (14, 15) have shown that gas resides in the polymer after foaming, the weights of foamed samples were measured after the samples were allowed to desorb the gas for at least 2 weeks.
An SEM (Hitachi Model S-2500) analysis was performed to characterize the foam samples. The samples were frozen in liquid nitrogen and fractured to ensure that the microstructure remained clean and intact, and then gold coated using a sputter coater (E 5000C PS3) for enhanced conductivity.
The bubble diameter, the void fraction ([V.sub.f]) and the cell-population density per unit volume of the original unfoamed polymer ([N.sub.o]) were characterized from the SEM micrographs using the following equations (28, 29):
[V.sub.f] = 1 - [[Rho].sub.F]/[Rho] (2)
[N.sub.o] = [([nM.sup.2]/A).sup.3/2] [1/1 - [V.sub.f]] (3)
where n is the number of bubbles in the micrograph, and A and M are the area and the magnification factor of the micrograph, respectively.
Mechanical Property Testing
Tensile testing was carried out according to ASTM standard D638 on Instron Universal Testing Machine Sintech (Model 20) equipped with Testwork program (Version 2.10, Sintech Inc.) for statistical calculations. The crosshead displacement rate was 6.5 mm/min. Three tensile properties were determined from the load versus elongation curves: (a) stress at break, (b) modulus, and (c) elongation at break. These properties were also expressed as specific properties, i.e., the ratio of each property over the density of material. The tensile strength was determined as the break load divided by the initial section area of the specimen. The initial tangent modulus was determined from the slope of the load-deformation curves. The deformations were measured by a strain gauge extensometer (Sintech, Model 632) and the elongation at break of a tensile test bar was recorded as percent elongation. All evaluations were made at room temperature (21 [degrees] C). In order to remove the effect of the residual gas on the properties, a sufficient time of at least two weeks was allowed for the gas to diffuse out of the samples after foaming. Before testing, the samples were conditioned at room temperature and 50% relative humidity for more than 48 hours in accordance with Procedure A of method ASTM D618. Five specimens were tested at each foaming condition.
The impact strengths of unfoamed and foamed samples were determined using Tinius Olsen Izod impact tester. Notched samples were tested at room temperature according to the procedure outlined in ASTM D256. All the samples were degassed before the testing as in the other property characterization to remove the effect of the residual gas.
RESULTS AND DISCUSSION
The dependence of the tensile and impact properties of microcellular foamed PVC/wood-fiber composites on the cell morphology were investigated in order to establish the structure-property relationships for these materials. The cell morphologies of foamed PVC/wood-fiber composites were studied first with respect to the foaming temperature and the surface treatment of fibers. Second, the tensile and impact properties were characterized in terms of the void fraction of foamed samples. For comparison, the cell morphology and mechanical properties of microcellular foamed PVC without any wood-fibers that were processed and characterized under the same conditions (12) are illustrated in the Figures together with those of microcellular foamed PVC/wood-fiber composites.
Dependence of Cell Morphology on the Foaming Temperature
Figure 1 illustrates the effect of foaming temperature on the void fraction for microcellular foamed PVC and PVC/wood-fiber composites with untreated and silane-treated fibers. For the PVC/wood-fiber composites with treated fibers, the void fraction increased as the foaming temperature increased as in the case of unfilled PVC. For the composites with untreated fibers, a little increase in the void fraction was observed up to 90 [degrees] C, and the void fraction leveled off above 90 [degrees] C. In general, the void fraction was decreased by the addition of wood-fibers in the PVC matrix. However, the surface modification of fibers increased the void fraction in the composites.
The void fraction was determined by cell growth phenomena. Mechanisms of cell growth are primarily governed by both the foaming temperature and foaming time (22). The foaming temperature affects both the gas/polymer matrix stiffness and the gas diffusion rate (30). When the polymer/gas solution is heated above its glass transition temperature during the foaming process, the stiffness of the polymer matrix is lowered, the gas diffusion rate is increased, and the cells start growing. The cells continue to grow and reduce the total material density as the gas molecules diffuse into the nucleated cells from the polymer matrix.
Larger void fractions were obtained in the composites with treated fibers than in the composites produced with untreated fibers. However, the void fractions of treated wood-fiber composites were in general much lower than those of PVC. It seemed that the void fractions obtained in foamed samples were primarily governed by the C[O.sub.2] concentration in the materials, as shown in Fig. 2. The concentration of C[O.sub.2] in PVC was 9.3%. In the composites with treated and untreated fibers, the measured concentrations of C[O.sub.2] were 8.0% and 3.4%, respectively. The addition of wood-fibers in the PVC matrix reduced the concentration of C[O.sub.2] in the composites because wood-fibers do not absorb C[O.sub.2] (23). The measured concentration of C[O.sub.2] absorbed by the composites with untreated fibers was much lower because of the poor interface between fibers and PVC matrix (3, 22, 23). Since the poor interface can provide a channel through which the gas can quickly escape from the composite to the environment, the loss of gas during the pressure-release time period before the weight gain measurement or before and during microcellular foaming must have been accelerated (23). However, when treated fibers were used in the processing of composites, the enhanced interface must have prevented the fast gas escape during the pressure-release time period and during microcellular foaming. This was why the measured C[O.sub.2] concentration in the composites with treated fibers was much higher than that in the composites with untreated fibers, although it is believed that the actually dissolved concentrations of C[O.sub.2] in the composites with treated and untreated fibers were almost the same. Since the C[O.sub.2] loss before and during foaming was accelerated in the composites with untreated fibers, most of the gas dissolved in the polymer matrix was not effectively used for cell growth, and the final void fraction achieved by foaming was very low. In contrast, in the composites with treated fibers, the dissolved gas was well utilized for cell growth, and consequently, a cell structure of a relatively higher void fraction was developed in these materials (22).
The effect of foaming temperature on the cell-population density for PVC and composites with treated fibers is shown in Fig. 3a. The cell-population density of each material was not affected by the foaming conditions used in this study. A relatively smaller number of bubbles were nucleated in the composites compared to PVC. The cell-population densities of PVC/woodfiber composites were on the order of [10.sup.9] cells/[cm.sup.3], one order of magnitude smaller than that of PVC. The microcell nucleation mechanism is strongly influenced by the saturation pressure, and to a certain extent, by the saturation time (28, 29). Therefore, one would expect that changes in the saturation pressure would have a great influence on the microcell nucleation rate and the resultant cell-population density. However, since all the samples in this study were saturated at the same pressure, the observed cell densities were the same in each material. Also, Fig. 3a implies that cell coalescence did not occur actively during foaming (22). Figure 3b illustrates the effect of foaming temperature on the average cell size. The average cell size increased with the foaming temperature for both PVC and composites with treated fibers. Because the cell-population density was the same (Fig. 3a), the increased average cell size was due to the larger expansion of cells via gas diffusion into the nucleated cells through the softened PVC matrix at a higher temperature.
The microcellular foaming conditions were carefully selected from the previous studies on the processing-structure relationships of microcellular foamed PVC/wood-fiber composites (22) to achieve a desired cell morphology, i.e., a cell density and a void fraction. First, the cell density was controlled by the saturation pressure. Then, the foaming time for each foaming temperature was selected to achieve a desired void fraction in the foam sample while preventing cell coalescence. The following section presents the mechanical property characterization methods used on foamed samples to establish the structure-property relationships for these materials.
Effect of Microcellular Foaming on the Tensile Strength at Break
Figure 4a illustrates the dependence of the tensile strength at break on the void fraction for microcellular foamed PVC and PVC/wood-fiber composites. A linear decrease of strength at break was observed as the void fraction increased for all materials studied. For the composites with treated fibers, the strength at break decreased from 21.8 MPa to 8.4 MPa as the void fraction increased from 0 to 0.56, while with untreated fibers in the composites, the decrease in tensile strength at break was from 17.3 MPa to 9 MPa as the void fraction increased from 0 to 0.20. The plot of specific tensile strength at break versus the relative density ([[Rho].sub.F]/[Rho]), which is related to the void fraction by Eq 2, is shown in Fig. 4b. It appears that the tensile strength at break to weight ratio is not compromised by foaming.
Interestingly, the tensile strength at break of the unfoamed composites with untreated fibers was almost 30% lower than that of unfoamed PVC. It is believed that the decreased tensile strength was due to the poor interfacial adhesion between PVC and woodfibers in the composites. Similar results have been reported by others (4, 5, 7, 31). Unlike the composites with untreated fibers, the treatment of fibers with silane improved the tensile strength at break of unfoamed composites. When silane was added onto the fiber during mixing, a strong cohesive interaction between fiber, silane, and polymeric matrix was created: the polymer and the fibers were coupled together through a chemical bonding (3, 27). However, when the PVC and PVC/wood-fiber composites were microcellular foamed, the tensile strength at break of the foamed materials was mainly governed by the cellular structure, i.e., the void fraction as shown in Fig. 4a.
Effect of Microcellular Foaming on the Tensile Modulus
The dependence of the tensile modulus on the void fraction is illustrated in Fig. 5a. A linear decrease of stiffness was generally noticed as the density of the foamed samples decreased. For the composites with treated fibers, the tensile modulus decreased from 0.483 GPa to 0.118 GPa as the void fraction increased from 0 to 0.56.
The measured stiffness of the foamed composites with untreated fibers demonstrated significant scatter. This may be due to a nonuniform dispersion of wood-fibers in the PVC matrix during mixing when a coupling agent was not used (3, 32). When a coupling agent is not used during compounding, the fibers tend to agglomerate each other because of strong fiber-fiber interactions resulting from hydrogen bonding, and the dispersion of fibers into the matrix becomes nonuniform (32). Because of the nonuniform dispersion of wood-fibers, the developed foam structures were also very nonuniform and irregular. Consequently, the measured stiffness of the foamed composites was scattered.
The modulus of the unfoamed composite is almost five to six times higher than that of the unfoamed PVC. based on the rule of mixtures (33), the modulus of short-fibers composites ([E.sub.c]) is proportional to the fiber's modulus and can be estimated by
[E.sub.c] = [k.sub.eff] [multiplied by] [E.sub.f] [multiplied by] (1 - [V.sub.m]) + [E.sub.m][V.sub.m] (4)
where [V.sub.m] is the matrix volume fraction, [E.sub.f] and [E.sub.m] are the fiber and matrix moduli, respectively, and [k.sub.eff] is the efficiency coefficient that varies as a function of the composites' microstructural parameters such as dispersion, orientation, adhesion, etc. Chtourou et al. (6) determined experimentally that in compression molding (as in the case of this study), the orientation of the fibers was random in all directions and that [k.sub.eff] related to the modulus of compressed samples is approximately 0.33. During processing of wood-fiber composites, wood-fibers tend to collapse into ribbons, and these ribbons possess higher stiffness in the range of 10 to 80 GPa (34). The higher stiffness of wood-fiber, when incorporated into the polymer matrix, increases the modulus of the wood-fiber composites so long as interface debonding does not occur.
The surface treatment of fibers had an almost negligible effect on the stiffness of the unfoamed composites, since the composites with both untreated and treated fibers had almost the same tensile modulus. These results indicated that the moduli of composites produced in this study were not sensitive to the adhesion between the polymer and the wood-fibers when testing stress levels were below the critical debonding stress. However, when these composites were foamed, the resultant stiffness was significantly different because the surface treatment affected the foam morphology. It may also be noted that although the tensile modulus of the composites was affected by foaming, foamed composites with a void fraction up to 56% were still stiffer than the unfoamed PVC, owing to the presence of wood-fibers.
When the data was expressed as specific modulus [ILLUSTRATION FOR FIGURE 5 OMITTED], a good correlation between the specific modulus and the relative density could not be observed. Since it is known that the modulus of polymeric foam with closed cells can be described in terms of the relative foam density as (35-38)
[E.sub.F]/E [approximately equal to] [([[Rho].sub.f]/[Rho]).sup.2] = [(1 - [V.sub.f]).sup.2] (5)
the experimental results were plotted together with the theoretically predicted values (based on Eq 5) in Fig. 5c. The stiffness of the microcellular foamed PVC/wood-fiber composites with treated fibers fitted well to theoretical values, whereas the stiffness of microcellular foamed PVC exhibited higher values than the predicted ones (12).
Effect of Microcellular Foaming on the Elongation at Break
In Fig. 6a, the elongation at break is plotted versus the void fraction for the PVC and wood-fiber composites. For the PVC/wood-fiber composites with treated fibers, the elongation at break was around 36% all the time regardless of the void fraction whereas a linear decrease of elongation at break from 115% to 29% was observed for PVC as the void fraction increased from 0 to 0.82 (12). However, no obvious trend was observed for the foamed PVC/wood-fiber composites with untreated fibers. Again, we believe that this data scatter was due to the inconsistent cellular structures developed in the composite with untreated fibers.
It may be noted that the elongation at break of the PVC/wood-fiber composites was much lower than that of PVC. The decreased elongation at break may be due to the mode of failure of each material. Our previous study showed that PVC/wood-fiber composites exhibits a brittle failure while unfilled PVC exhibits a more ductile failure (31). It seemed that the ductile mode of failure of the polymer matrix was altered by the incorporation of the wood-fibers. In other words, the elongation at break was sacrificed by introducing the brittle wood-fibers into the polymer matrix. Another interesting result was that the elongation at break of the unfoamed composite with treated fibers was lower than that of the unfoamed composite with untreated fibers. The creation of a strong cohesive interaction between fibers and polymer is believed to increase the brittleness of the composites. While surface treatment led to an improvement in tensile strength, it caused a marked reduction in elongation at break. Similar results have been reported by other investigators (7-9). Nevertheless, the surface treatment of wood-fibers seems to be beneficial because the foamability is enhanced (22, 23) and the developed foam structure is favorable for enhancing the ductility of the materials.
The dependence of the specific elongation at break on the relative density of microcellular foamed PVC and PVC/wood-fiber composites is shown in Fig. 6b. The specific elongation at break increased as the relative density decreased. Two distinct trends of specific elongation at break were observed for the PVC and the PVC/wood-fiber composites with treated fibers. The presence of wood-fibers decreased the specific elongation at break for all relative densities.
Effect of Microcellular Foaming on the Notched Izod Impact Strength
The concept of improving the impact strength of PVC/wood-fiber composites by introducing a microcellular structure in the composite matrix was examined. The measured notched Izod impact strengths of microcellular foamed PVC/wood-fiber composites are shown in Fig. 7. An improvement in the impact strength of PVC/wood-fiber composites with treated fibers was observed after foaming. The notched Izod impact strengths of foamed composite samples were, in general, two to three times larger than those of the unfoamed samples depending on the void fraction. However, no discernible improvement in impact strength was observed by foaming the composites with untreated fibers.
It should be noted that the notched Izod impact strengths of unfoamed PVC/wood-fiber composites were much lower than that of unfoamed PVC irrespective of the fiber surface treatment. The average notched Izod impact strength of unfoamed PVC was measured to be 71.5 J/m (12). The average notched Izod impact strengths of unfoamed PVC/wood-fiber composites with treated and untreated fibers were 30.5 [+ or -] 3.2 J/m and 37.9 [+ or -] 2.5 J/m, respectively. As in the case of the elongation at break, the observed notched Izod impact strength of unfoamed PVC/wood-fiber composites with treated fibers was slightly lower than that of unfoamed composites with untreated fibers. This observed phenomenon may be attributed to the higher interfacial adhesion of the treated fiber composites than that of the untreated fiber composites. The enhanced interaction between fibers and polymer is believed to increase the brittleness of the composites by changing the mode of failure from "fiber pull-out" to "fiber breakage" when a crack propagates (39). When untreated wood-fibers are incorporated into the PVC matrix, the lack of intimate adhesion between fibers and polymer causes the fibers to be pulled out and a larger amount of energy is required during crack propagation. On the other hand, when the interface is improved by using treated fibers in the composites, the crack goes through the brittle wood-fiber, and as a result, the overall composite materials become more brittle and the impact strength of the material is reduced. Similar results have been reported in other work (7, 40).
When these materials were microcellular foamed, the notched Izod impact strengths were improved for all the materials as the void fraction increased. For example, when the PVC/wood-fiber composites with treated fibers were foamed with a void fraction of 0.56, the improvement of impact strength was about three times as high as that of the unfoamed composites. For the microcellular foamed PVC with a void fraction of 0.80, the notched Izod impact strength was about four times the unfoamed PVC on the average (12). However, the PVC/wood-fiber composites with untreated fibers did not foam well, and as a consequence, the notched Izod impact strength did not improve much. Since the degree of notched Izod impact strength improvement was a strong function of the void fraction, a correlation exists between foamability and notched Izod impact strength. Figure 8 illustrates this effect of foamability on the notched Izod impact strengths of PVC and PVC/wood-fiber composites with treated and untreated fibers. The data shown in Fig. 8 were collected from samples of microcellular foams produced at 90 [degrees] C for 5 seconds. The void fractions of the foamed PVC, the foamed composites with treated fibers, and the foamed composites with untreated fibers were 0.75, 0.56, and 0.20, respectively. The well-foamed samples of PVC and PVC/wood-fiber composites with treated fibers exhibited improved notched Izod impact strength whereas the composites with untreated fibers showed a negligible change because of the poorly developed foam structure.
It should be emphasized that the deteriorated notched Izod impact strength of wood-fiber composites due to the incorporation of brittle wood-fibers was well recovered by the introduction of a microcellular structure to the composites. This increase in notched Izod impact strength may be attributed to the presence of fully grown small bubbles which inhibit crack propagation by blunting the crack tip and increasing the amount of energy needed to propagate the crack (12, 13). It is also believed that the struts and cell walls of the foam structure absorbed the energy during deformation, and thereby increased the impact strength. However, if the fibers are not treated properly, the plastic/wood-fiber composites may not be able to take advantage of microcellular foaming to improve the impact strength because of the poor foamability.
SUMMARY AND CONCLUSIONS
This study examined the effects of the void fraction and the fiber surface treatment on the tensile and impact properties of foamed PVC/wood-fiber composites. Experiments were carried out in two parts to fulfill this objective. First, microcellular foamed structures were developed in PVC/wood-fiber composites by saturating the samples with C[O.sub.2], followed by rapidly decreasing the solubility of gas in the samples. The void fraction of microcellular foamed samples was controlled by tailoring the composition of materials and the foaming temperature. Second, tensile and impact tests were conducted on the foamed samples to investigate the dependence of these properties on the cell morphology of microcellular foamed materials.
The main motivation for this study was to investigate the effects of microcellular structures developed in PVC/wood-fiber composites on the mechanical and physical properties. Since a microcellular foam structure reduces the weight and improves the impact strength of the materials, the shortcomings of plastic/wood-fiber composites, i.e., the deteriorated impact strength and the high density, were improved by incorporating a microcellular foam structure in the PVC/wood-fiber composites.
Based on the experimental results, the following conclusions can be drawn:
1. For the unfoamed materials, the addition of wood-fibers in the PVC matrix reduced the tensile strength at break of the composites because of a poor interface between fibers and polymer. Enhancing the adhesion of fibers to polymer has been found to be a suitable means for improving composite's strength.
2. The tensile strength at break of PVC/wood-fiber composites was deteriorated by foaming. It was observed that the tensile strength decreased as the void fraction increased. However, the specific tensile strength was maintained regardless of the void fraction.
3. The tensile moduli of unfoamed PVC/wood-fiber composites were higher than the neat PVC due to the higher modulus of wood-fiber filler. The surface treatment of the fibers did not affect the tensile modulus of unfoamed PVC/wood-fiber composites significantly.
4. The tensile modulus of PVC/wood-fiber composites was deteriorated by foaming. The tensile modulus of foamed composites with treated fibers decreased with an increase in the void fraction and the deterioration followed the predicted values. However, the tensile moduli of foamed composites with untreated fibers were observed to be very inconsistent because of the poorly dispersed fibers in the PVC matrix and poorly developed foam structures.
5. The elongation at break of unfoamed PVC/woodfiber composites was much lower than that of the unfoamed PVC. The enhancement of the fiber-PVC interface did not improve this property but made the composites behave more brittle.
6. The microcellular foaming of the PVC/woodfiber composites alleviated the deterioration of the elongation at break. The specific elongation at break increased as the void fraction increased. At all void fractions, the PVC/woodfiber composite foams exhibited less elongation at break than their PVC counterpart foams of the same void fraction.
7. The impact strengths of unfoamed PVC/woodfiber composites were much lower than that of the unfoamed PVC. The impact strengths of the composites with treated fibers were worse than that with untreated fibers as in the case of elongation at break.
8. The deteriorated notched Izod impact strength of PVC/wood-fiber composites was improved by microcellular foaming. The notched Izod impact strength of foamed composites increased as the void fraction increased. Since a microcellular foamed structure was not developed well in the composites with untreated fibers, the improvement of notched Izod impact strength was greater when the treated fibers were used in composite processing.
The authors are grateful for the generous donation of materials and financial support provided by Royal Plastics Limited. The contribution of MRCO for the equipment used in this study is also acknowledged. Le Fonds pour la Formation de Chercheurs et l'Aide a la Recherche (Fonds FCAR) of Quebec and the University of Toronto are also acknowledged for the scholarships awarded to one of the authors (L.M.M.). Last, the authors would like to thank Mr. S. Law for his useful technical advice during manufacturing and property testing of composites.
A = Area of the micrograph, [cm.sup.2].
E = Modulus of unfoamed samples, GPa.
[E.sub.c] = Composite's modulus, GPa.
[E.sub.f] = Fiber's modulus, GPa.
[E.sub.F] = Modulus of foamed samples, GPa.
[E.sub.m] = Modulus of the matrix, GPa.
[K.sub.eff] = Efficiency coefficient.
M = Magnification factor for the micrograph.
[M.sub.a] = Weight of the sample measured in air, g.
[M.sub.w] = Weight of the sample measured in water, g.
[N.sub.o] = Cell-population density, cells/[cm.sup.3].
N = Number of bubbles in the SEM micrograph, cells.
[Rho] = Density of unfoamed material, g/[cm.sup.3].
[[Rho].sub.F] = Density of foam, g/[cm.sup.3].
[V.sub.f] = Void fraction.
[V.sub.m] = Matrix volume fraction.
1. J. J. Balatinecz and R. T. Woodhams, J. Forestry, 91, 21 (1993).
2. J. A. Youngquist and R. M. Rowell, in 23rd International Particleboard/Composite Materials Symposium Proceedings, p. 141, T. M. Maloney, ed., Washington. State Univ., Pullman, Wash. (1990).
3. L. M. Matuana, PhD thesis, Faculty of Forestry, University of Toronto, Toronto, Ontario, Canada (1997).
4. B. V. Kokta, D. Maldas, C. Daneault, and P. Beland, J. Vinyl Technol, 12, 146 (1990).
5. D. Maldas, B. V. Kokta, and C. Daneault, J. Vinyl Technol., 11, 90 (1989).
6. H. Chtourou, B. Riedl, and A. Ait-Kadi, J. Reinf. Plast. Compos., 11, 372 (1992).
7. M. Y. A. Fuad, Rahmad Shukor, Z. A. Mohd Ishak, and A. K. Mohd Omar, Plast. Rubber Proc. Appl., 21, 225 (1994).
8. R. G. Raj, B. V. Kokta, G. Groleau, and C. Daneault, Plast. Rubber Proc. Appl., 11, 215 (1989).
9. R. G. Raj, B. V. Kokta, and J. D. Nizio, J. Appl. Polym. Sci., 45, 91 (1992).
10. J. Martini, F. A. Waldman, and N. P. Suh, SPE ANTEC Tech. Papers, 28, 674 (1982).
11. F. A. Waldman, SM thesis, Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Mass. (1982).
12. L. M. Matuana, C. B. Park, and J. J. Balatinecz, Cell. Polym. 17 (1998).
13. S. Doroudiani, C. B. Park, and M. T. Kortschot, Polym. Eng. Sci., 38, 1205 (1998).
14. D. I. Collias and D. G. Baird, Polym. Eng. Sci., 35, 1167 (1995).
15. D. I. Collias and D. G. Baird, Polym. Eng. Sci., 35, 1178 (1995).
16. D. I. Collias, D. G. Baird, and R. J. M. Borggreve, Polymer, 25, 3978 (1994).
17. D. F. Baldwin and N. P. Suh, SPE ANTEC Tech. Papers, 38, 1503 (1992).
18. K. A. Seeler and V. Kumar, J. Reinf. Plast. Compos., 12, 359 (1993).
19. K. A. Seeler and V. Kumar, SPE ANTEC Tech. Papers, 40, 1972 (1994).
20. M. Shimbo, D. F. Baldwin, and N. P. Suh, Polym. Eng. Sci., 35, 1387 (1995).
21. V. Kumar and N. P. Suh, SPE ANTEC Tech. Papers, 34, 715 (1988).
22. L. M. Matuana, C. B. Park, and J. J. Balatinecz, Polym. Eng. Sci., 37, 1137 (1997).
23. L. Matuana-Malanda, C. B. Park, and J. J. Balatinecz, J. Cell. Plast., 32, 449 (1996).
24. V. Kumar and J. E. Weller, Intern. Polym. Proc., 8, 73 (1993).
25. M. R. Holl, V. Kumar, M. Ma, and R. R. Kwapisz, Jr., SPE ANTEC Tech. Papers, 42, 1908 (1996).
26. J. E. Weller, M. Ma, R. Montecillo, and V. Kumar, SPE ANTEC Tech. Papers, 43, 841 (1997).
27. L. M. Matuana, R. T. Woodhams, J. J. Balatinecz, and C. B. Park, Polym. Compos., 19, 446 (1998).
28. D. F. Baldwin, C. B. Park, and N. P. Suh, Polym. Eng. Sci., 36, 1437 (1996).
29. D. F. Baldwin, C. B. Park, and N. P. Suh, Polym. Eng. Sci., 86, 1446 (1996).
30. D. F. Baldwin, N. P. Suh, and M. Shimbo, in Cellular Polymers, V. Kumar and S. Advani, eds., p. 109, MD-Vol. 38, ASME, New York (1992).
31. L. M. Matuana, C. B. Park, and J. J. Balatinecz, J. Vinyl & Additive Technol., 3, 265 (1997).
32. J. Felix, PhD thesis, Department of Polymer Technology, Chalmers University of Technology, Goteborg, Sweden (1993).
33. D. Hull, An Introduction to Composite Materials, Cam-bridge University Press, New York (1981).
34. R. T. Woodhams, G. Thomas, and D. K. Rodgers, Polym. Eng. Sci., 24, 1166 (1984).
35. M. F. Ashby and D. R. H. Jones, Engineering Material 2: An Introduction to Microstructures, Processing and Design, Pergamon Press, Toronto (1986).
36. L. J. Gibson and M. F. Ashby, Cellular Solids: Structure and Properties, Pergamon Press, New York (1988).
37. D. R. Moore, K. H. Couzens, and M. J. Iremonger, J. Cell. Plast., 10, 135 (1974).
38. R. C. Progelhof and J. L. Throne, Polym. Eng. Sci., 19, 493 (1979).
39. G. A. Cooper and M. R. Piggot, "Cracking and Fracture in Composites," Proc. 4th International Conference on Fracture, Vol. 1, pp. 557-605, Waterloo, Canada, (June, 1997).
40. J. K. Kim and Y. W. Mai, Compos. Sci. Technol., 41, 333 (1991).
|Printer friendly Cite/link Email Feedback|
|Author:||Matuana, Laurent M.; Park, Chul B.; Balatinecz, John J.|
|Publication:||Polymer Engineering and Science|
|Date:||Nov 1, 1998|
|Previous Article:||Solid state and mechanical properties of injection molded poly(ether ether ketone)/polyarylate blends.|
|Next Article:||Physical and relaxation properties of flame-sprayed ethylene-methacrylic acid copolymer.|